Localizing Objects with Self-Supervised Transformers and no Labels

Localizing objects in image collections without supervision can help to avoid expensive annotation campaigns. We propose a simple approach to this problem, that leverages the activation features of a vision transformer pre-trained in a self-supervised manner. Our method, LOST, does not require any external object proposal nor any exploration of the image collection; it operates on a single image. Yet, we outperform state-of-the-art object discovery methods by up to 8 CorLoc points on PASCAL VOC 2012. We also show that training a class-agnostic detector on the discovered objects boosts results by another 7 points. Moreover, we show promising results on the unsupervised object discovery task. The code to reproduce our results can be found at https://github.com/valeoai/LOST.

PDF Abstract

Results from the Paper

Ranked #4 on Weakly-Supervised Object Localization on CUB-200-2011 (Top-1 Localization Accuracy metric)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Single-object discovery COCO_20k LOST + CAD CorLoc 57.5 # 6
Single-object discovery COCO_20k LOST CorLoc 50.7 # 8
Weakly-Supervised Object Localization CUB-200-2011 LOST Top-1 Localization Accuracy 71.3 # 4